# US 11,908,116 B2 · Scaled PSNR for image quality assessment

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,908,116 B2
- Document ID: US11908116B2
- Status: Granted
- Kind: B2
- Application number: 17/222,645
- Filed: 2021-04-05
- Published: 2022-02-17
- Granted: 2024-02-20
- Earliest claimed: 2020-08-11
- Assignees: Sony Group Corporation
- Applicants: Sony Group Corporation
- Inventors: Jong Hwa Lee; Praggya Garg
- Web page: https://patentgrep.com/patent/US11908116B2

## CPC

- G06T 7/0002: {Inspection of images, e.g. flaw detection}
- G06V 10/751: {Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching}
- G06V 10/761: {Proximity, similarity or dissimilarity measures}
- G06V 10/993: {Evaluation of the quality of the acquired pattern}
- G06T 2207/30168: Image quality inspection
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

The scaled Peak Signal-to-Noise Ratio (PSNR) method scales PSNR to 1.00 or 100% from infinite decibel values. The PSNR describes the ratio between the maximum possible value (power) of a signal and the power of signal's noise. Noise affects the quality of its representation. Signals dynamically fluctuate from the smallest to largest possible quantity. The unsettled range to maximum from minimum values of the noise is unknown. Therefore, the PSNR usually forms on a logarithmic way (log) of describing a ratio, the decibel (dB). The dB is able to gauge the level of power, communication signal (intensity or voltage), sound pressure, and intensity of several other items including images.

## Claims

### Claim 1

A method programmed in a non-transitory memory of a device, the method comprising:
acquiring a first image and a second image;
comparing the first image and the second image using a scaled Peak Signal-to-Noise Ratio (sPSNR) to generate an image quality assessment value;
comparing the image quality assessment value with a threshold, wherein the sPSNR is utilized for computer vision for machine learning, wherein the sPSNR is based on the equation:
sPSNR
=
10
⁢
log
⁡
(
c
0
⁢
MAX
4
1
+
MSE
)
.
wherein c0 is a scale factor, MAX is a maximum possible pixel value, and MSE is a mean squared error;
comparing luminance, contrast and one or more structures in the first image and the second image to generate a structural similarity index; and
combining the sPSNR and the structural similarity index to determine whether the first image and the second image match.

### Claim 2 (depends on 1)

The method of claim 1 wherein the sPSNR is based on a scale factor, a maximum possible pixel value and a mean squared error.

### Claim 3 (depends on 1)

The method of claim 1 wherein the image quality assessment value comprises a percentage value between 0% and 100%, inclusive.

### Claim 4 (depends on 1)

The method of claim 1 wherein the threshold is generated by a user and/or machine learning.

### Claim 5

An apparatus comprising:
a non-transitory memory for storing an application, the application for:
acquiring a first image and a second image;
comparing the first image and the second image using a scaled Peak Signal-to-Noise Ratio (sPSNR) to generate an image quality assessment value;
comparing the image quality assessment value with a threshold, wherein the sPSNR is utilized for computer vision for machine learning, wherein the sPSNR is based on the equation:
sPSNR
=
10
⁢
log
⁡
(
c
0
⁢
MAX
4
1
+
MSE
)
.
wherein c0 is a scale factor, MAX is a maximum possible pixel value, and MSE is a mean squared error;
comparing luminance, contrast and one or more structures in the first image and the second image to generate a structural similarity index; and
combining the sPSNR and the structural similarity index to determine whether the first image and the second image match; and
a processor coupled to the memory, the processor configured for processing the application.

### Claim 6 (depends on 5)

The apparatus of claim 5 wherein the sPSNR is based on a scale factor, a maximum possible pixel value and a mean squared error.

### Claim 7 (depends on 5)

The apparatus of claim 5 wherein the image quality assessment value comprises a percentage value between 0% and 100%, inclusive.

### Claim 8 (depends on 5)

The apparatus of claim 5 wherein the threshold is generated by a user and/or machine learning.

### Claim 9

A system comprising:
a first device configured for acquiring a first image and a second image; and
a second device comprising a non-transitory memory for storing an application, the application configured for:
comparing the first image and the second image using a scaled Peak Signal-to-Noise Ratio (sPSNR) to generate an image quality assessment value;
comparing the image quality assessment value with a threshold, wherein the sPSNR is utilized for computer vision for machine learning, wherein the sPSNR is based on the equation:
sPSNR
=
10
⁢
log
(
c
0
⁢
MAX
4
1
+
MSE
)
,
wherein c0 is a scale factor, MAX is a maximum possible pixel value, and MSE is a mean squared error;
comparing luminance, contrast and one or more structures in the first image and the second image to generate a structural similarity index; and
combining the sPSNR and the structural similarity index to determine whether the first image and the second image match.

### Claim 10 (depends on 9)

The system of claim 9 wherein the sPSNR is based on a scale factor, a maximum possible pixel value and a mean squared error.

### Claim 11 (depends on 9)

The system of claim 9 wherein the image quality assessment value comprises a percentage value between 0% and 100%, inclusive.

### Claim 12 (depends on 9)

The system of claim 9 wherein the threshold is generated by a user and/or machine learning.